Objectives of this proposal are to determine predictors of needlestick injuries in nursing personnel in relation to use of safety products and to utilize this data to prevent needlestick injuries in the future. Specific aims: Phase I: Examine the relationship between sociodemographic and work-related characteristics of nurses and use of safety devices to prevent needlesticks. It will consist of a cross-sectional survey of 2400 nursing personnel at a medical center in San Jose, California. Estimated duration is 12 months (survey development, pilot testing, distribution, recontact of nonresponders, data analysis, and write-up of findings). Statistical methodology: cross tabulations and correlations between characteristics of nurses and use of safety devices; Logistic regression to examine the relationship between independent variables (sociodemographic and work-related) and the dependent variable, use of safety devices. Phase II: Examine the characteristics of the percutaneous injuries that occur among nurses, with the ultimate goal of determining predictors of needlestick injury, and will consist of a case control study of nurses with needlestick injuries (n=50) and a suitable comparison group of nurses who did not have needlestick injuries (n=200). Statistical methodology: Descriptive statistics to characterize needlestick injuries in cases; Logistic regression to identify predictors of needlestick injuries. Phase II will be completed in two years for questionnaire design, pilot and revision; data collection for cases and controls; data entry and analysis; and preparation of dissertation.